Cognition

This chapter considers the fundamental characteristics of human cognition. It suggests that the capacity to make sense of the world involves a complex interaction between external realities and mental models. It considers that there are two complementary types of model—holistic representations, and models consisting of abstract componentsorganized in categories and hierarchies. It suggests that knowledge can be described as stable sets of models and their connections. It notes that reasoning is usually described as having two modes: Intuition operates rapidly and unconsciously; analysis requires sustained conscious effort. It argues that an absolute distinction between the two modes is artificial; intuition is usually the driving force, but effective reasoning depends on the synergy of both modes. It introduces the concept of bias and suggests that bias and intuition are inseparable.

Safeguards

This chapter refers to the 2015 report Improving Diagnosis in Health Care by the Institute of Medicine (IOM), which sets out the case for changes in healthcare organizations, professional education, and clinical practice. It notes that the insights in the IOM report are not new, but the pace of reform has been slow. It suggests that both the inertia of large organizations and professional and institutional resistance to change are implicated in this delay. It suggests that, despite methodological challenges, there is evidence that changes in clinical processes and professional reflective practice can improve diagnostic accuracy. It reiterates the importance of clinical skills, suggests modifications to medical record-keeping, proposes enhanced procedures for generating and verifying the differential diagnosis, and suggests that computerized diagnostic decision support has potential. It concludes that evidence and expert opinion support these changes, but acknowledges there will be a battle for hearts and minds.

Probability in Diagnostic Thinking

This chapter observes that diagnostic reasoning involves both informal and mathematical estimates of probability. It argues that intuitive estimates of the likelihood of disease are necessary in the early phases of the diagnostic process but notoriously inaccurate. It notes that formal calculations are not possible when the question is, What might be wrong with this person? but are much more accurate than intuition in estimating the probability that a specific disease is present. The chapter suggests that population-based calculations of the likelihood of disease may lead clinicians to play Russian roulette by proxy because individual variation and individual risk factors may alter that risk in a given patient. It refers to evidence that many clinicians are inexpert in statistical methods. The chapter describes some basic statistical processes and their place in the clinical application of test results. It discusses the necessity and challenges of managing patients whose symptoms are medically unexplained.

Diagnosis Terms

This chapter considers diagnosis in its sense as a term within formal classifications. It argues that classifications are indispensable and work best for conditions that can be described precisely and objectively using the biomedical model of disease. It argues that diagnostic terms may have hidden layers of meaning, reflecting individual or cultural beliefs. It suggests that these unspoken meanings influence the outcome of treatment, and that harmful misunderstandings are not unusual. It argues that classifications are less clear when disease is defined by a threshold value on a spectrum consisting of phenomena that are also experienced by people who feel well, and that this difficulty can contribute to overdiagnosis. It argues that, in these circumstances, it is of paramount importance that doctors should inform and involve patients in decisions about treatment. It recognizes the value of the patient-centred approach but argues that this does not resolve all the dilemmas.

Triumphs, Disasters, and Near Misses

This chapter refers to the evidence that error occurs in about 10% of diagnoses and is a potent source of harm to patients. Most error is judged to be potentially avoidable and to result from defects in the thinking of the individuals involved, defects in the systems within which they work, or more commonly from problems occurring simultaneously in both. It notes that systems factors and fallible cognition each contributes to this figure and often coexist. The chapter offers a set of clinical anecdotes that illustrate the cognitive processes described in Chapters 2 and 4, with examples of common pitfalls, including the influence of bias. The chapter includes critical reflection on these case studies, considers what lessons may be taken, and identifies opportunities for the introduction of safeguards.

Diagnostic Reasoning

This chapter describes the diagnostic process and provides illustrations from the author’s experience. It notes that the clinical encounter has parallels in hypothetico-deductive reasoning, sharing key elements but not the sequential structure. It refers to the consensus that the thinking of experienced clinicians seldom follows a simple sequence and is driven by intuition, augmented by analytical approaches. Suggested is that success depends on their synergistic function. It argues that the greatest challenge in diagnostic reasoning is the need to integrate imperfect information to synthesize a coherent conclusion. It notes that the cognitive mechanisms are complex, obscure, and prone to error. It describes common biases and argues that both cognitive and affective bias are potent sources of error. The chapter proposes that the model of diagnostic reasoning implicit in traditional medical education should be replaced by a more explicit and complex description incorporating these insights.

Introduction

This chapter briefly describes the author’s background in primary care and the origins of his interest in diagnostic reasoning and error. It introduces the key themes of the book: the central role of diagnosis in medicine; the nature of diagnostic terms and classifications; the characteristics of human cognition; diagnostic reasoning; common types of diagnostic error; the implications for patient safety; the potential for changes in education, postgraduate medical training, and clinical practice to reduce diagnostic error; and possible obstacles to the implementation of the proposed changes. Also addressed is actual clinical encounters and how they are addressed by new clinicians and experienced physicians.